Comparison of Multiple Regression and Configural Analysis Techniques for Developing Base Expectancy Tables

Abstract

This study compares two statistical techniques-multiple regression and configural analysis—used in developing parole prediction tables, according to their ability to (1) differentiate between offenders who violate parole and those who do not, (2) predict violators from among a new group of parolees, and (3) assist administrators and researchers.First, experience tables had to be developed and tested for prediction ability. Once their accuracy in predicting had been demonstrated, they could be used as base expectancies because they had the quality of being "expected." As such, they could be used as a yardstick to evaluate correctional programs' ability to reduce these "expected" violation rates.The two methods were applied to the same body of data and the results were compared. The data consist of Wisconsin adult male offenders paroled in 1954-57 and in 1958-59. All were followed up for two years while they were on parole. The first group was used to develop the experience tables; the second group was used to test prediction ability.

Statistical Risk Assessment: Old Problems and New Applications

“…In criminal justice applications, this potential unfortunately remains largely theoretical. Several have attempted to demonstrate the relative utility of different statistical approaches to criminal justice prediction problems (e.g., Babst, Gottfredson, & Ballard, 1968;Farrington, 1978;D. M. Gottfredson & Ballard, 1964;Simon, 1971Simon, , 1972Wilbanks & Hindelang, 1972), and potential advantages of different approaches have been discussed by Wilkins and MacNaughton-Smith (1964; see also M. R. Gottfredson & Gottfredson, 1980;S.…”

The Development of a Parolee Classification System Using Discriminant Analysis

“…These tables are generally predicated upon either configural or multiple regression analyses (Babst et al, 1968). Most researchers have expressed a preference for configural analysis because, as stated by Inciardi (1971:66): &dquo;It has indicated considerable effectiveness and requires no highly technical and abstract coefficients or weights.&dquo;' However, there is a major defect of configural analysis which is not inherent in multivariate techniques: Configural analysis, by not taking into consideration the intercorrelation among the explanatory variables, cannot reveal the relative importance of each explanatory variable, given the inclusion of the other variables in the model.…”

Methods of Evaluating Correctional Programs: A Review of the Literature

“…Methods used have included both multiple regression and configural analysis. Babst, Gottfredson and Ballard (1968), in comparing the two methods, found the results achieved were similar (i.e., led to similar base expectancy tables). Each has its advantages: multiple regression provides more flexibility in choosing cut off scores, but configural analysis provides more predictive-power for sub-groups of the sample.…”